Ten genes that code for cell adhesion molecules have been identified by gene array analysis that are up-regulated in ovarian carcinoma tissue samples and are relatively specific to ovarian carcinoma. The objective of this project is to identify one or more cell adhesion molecule in which: (a) altered gene expression corresponds to altered protein expression, (b) protein expression is associated with patient information, and (c) the biological relevance of the molecules can be defined with respect to ovarian carcinoma. These molecules may potentially serve as early detection biomarkers or for detection of recurrence of ovarian carcinoma. In addition, these molecules may prove useful in the design of novel therapeutic regimens for ovarian carcinoma. In Aim #1, the ten cell adhesion proteins will be localized by immunohistochemistry in tissue samples of ovarian carcinoma vs. other tissue types. Their molecular weight in tissue samples will be verified by Western immunoblotting or radioimmunoprecipitation. In Aim #2, blood and ascites samples from patients with ovarian carcinoma will be screened for the presence of the proteins by Western immunoblotting or radioimmunoprecipitation. The level of each protein in the patients' circulation will be quantitated by ELISA or radioimmunoassay. Statistical analysis will be used to identify the association between tissue distribution, protein expression, and circulating levels of the proteins, and tissue histology and patient outcome/information. In Aim #3, the levels of expression of the proteins on the surface and in the spent media of fresh ovarian carcinoma patient samples and established cell lines will be quantitated. The role of these cell adhesion proteins in ovarian carcinoma cell adhesion, invasion, and aggregate formation will be elucidated by use of blocking monoclonal antibodies and RNA interference. The association between levels of expression/secretion of the cell adhesion proteins with biological functional activities and patient information will be determined.